kernels: MXFP4 -> BF16 dequant kernel (verified vs numpy)

From-scratch CUDA kernel for gpt-oss expert weights: one thread per packed
byte decodes 2 FP4 (E2M1) codes, applies the per-32-block E8M0 scale
(2^(e-127)), and writes BF16 transposed into [IN, OUT] (IN = nblk*32) so it
drops straight into x @ W. dequant_mxfp4() wrapper takes raw GpuBuffers
(uint8 is not an xserv Tensor dtype). mxfp4-check bin dequants layer-0
expert-0 on GPU and matches tools/mxfp4_probe.py exactly:
  [0, 0, 0, -0.0625, 0, -0, -0.015625, -0.03125]

This lets experts stay MXFP4-resident on GPU (13GB, fits one 32GB card)
and be dequantized to a BF16 scratch right before each expert GEMM, instead
of holding 36GB of BF16 or uploading experts per token. Loader plumbing +
GPU MoE decode use it next.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

crates/xserv-kernels/src/lib.rs

@@ -15,6 +15,7 @@ pub use attention::{attention, decode_attention, flash_attention, paged_decode_a
 pub use embedding::embedding;
 pub use gemm::{batched_matmul, matmul, GemmBackend};
 pub use layernorm::layernorm;
+pub use quant::dequant_mxfp4;
 pub use rmsnorm::{add_rmsnorm, rmsnorm};
 pub use rope::{rope_inplace, RopeCache};
 pub use softmax::softmax;